Introduction The selection of relevant sensory signals and generation of appropriate actions are fundamental functions of the primate brain. Disruptions of these functions are implicated in a variety of disorders, including attention deficit hyperactivity disorder (ADHD) and autism. Scientists in my section investigate the neuronal circuits involved in this visual function using a range of techniques, in both non-human primates and mice, in order to understand how these neuronal circuits operate under normal conditions and to identify how breakdowns in these mechanisms cause disorders of sensory-motor coordination. Standard models of visual attention emphasize the role of the cerebral cortex. In contrast, our results demonstrate that higher-order functions like selective attention are built on top of conserved subcortical circuits in the superior colliculus (SC), thalamus and basal ganglia, that play a central role in action selection. Our aims are to understand the operation of the subcortical structures and how they interact with the cortex, with the long-term goal of identifying the detailed neuronal circuit so that more specific therapeutic interventions can be developed. 1) Neuronal circuits for the control of selective attention in primates Most of our work is done using non-human primates, whose close homology with humans makes them the best animal model of human visual attention. 1a) Identification of a putative macaque homolog to temporal cortical structures known to play a central role in human neglect Spatial neglect is a brain dysfunction characterized by failures to orient, detect or respond to stimuli located in the visual field opposite the brain lesion. Spatial neglect is a common consequence of stroke, and involves damage to particular regions of cerebral cortex, but animal models are not yet established. In particular, the absence of evidence for temporal cortex involvement in macaque monkeys has suggested a fundamental difference from humans. We adopted a novel approach for identifying cortical structures in macaque that might be involved in neglect. It is now firmly established that unilateral inactivation of the mindbrain SC causes neglect-like deficits in macaque monkeys. We reasoned that functional imaging of the cerebral cortex combined with manipulations of SC activity might reveal cortical areas functionally related to neglect. We therefore combined reversible inactivation of the SC with fMRI as monkeys performed a demanding visual attention task and identified cortical regions whose attention-related modulation was reduced during the neglect-like deficits cause by suppression of SC activity. We found the largest reductions in fMRI attention-related modulation in a relatively unknown cortical region on the floor of the superior temporal sulcus. We also found smaller reductions in cortical regions that might have been expected, such as the prefrontal and dorsal parietal cortex. Additional experiments verified the importance of this cortical region. First, we repeated these experiments with inactivation of prefrontal cortex, and found loss of attention-related modulation in the same cortical region. Second, we repeated the experiments using a task that required attention to a different visual feature and verified that the loss of activation did not depend on the particular visual stimulus. Finally, we found that direct inactivation of this cortical region itself caused neglect-like deficits in spatial attention, similar to those previously shown for inactivation of the SC. These results provide the first demonstration of a causal link between neglect-like deficits and an attention-related region in the mid-STS cortex of monkeys. The cortical circuits for attention in monkeys therefore include an important node in the temporal cortex, which could provide an important animal model for understanding the etiology of neglect in humans. These results were published in March 2019 in Current Biology. 1b) Selective attention and the primate basal ganglia The basal ganglia are another set of subcortical and highly conserved brain structures that have been recently been implicated in selective attention. Last year we published the first demonstration of attention-related modulation in the caudate nucleus (the major input structure of the basal ganglia) of the primate, which was published in PloS Biology (Arcizet et al, 2018). In particular, we showed that activity in the caudate was sufficient to identify the different temporal epochs of an attention task. We have now built on these findings by testing how activity in caudate might be functionally linked to the superior colliculus. Our specific hypothesis was that outputs from the SC to the caudate nucleus (through the thalamus) are at least partly responsible for the spatial selectivity found on caudate neurons during attention tasks. We tested this idea by recording extracellular activity from populations of caudate neurons before and during unilateral inactivation of the SC in monkeys as they performed a covert attention task. Consistent with the hypothesis, we found that during inactivation many fewer caudate neurons showed attention-related modulation when the spatial cue was presented in the affected hemifield. In addition, we found that the ability of a machine learning classifier to correctly identify the task epoch, based on caudate neuron activity, was markedly reduced during SC inactivation. These results demonstrate that the role of the caudate nucleus in attention depends on signals it receives from the superior colliculus, and that these signals have a spatially specific effect on caudate processing. More broadly, these findings show the importance of subcortical loops from the superior colliculus to the basal ganglia as part of the circuits that support the function of visual selective attention. 2) Role of subcortical neuronal circuits in visual detection and attention in mice Mice provide opportunities to work out the details of neuronal circuits in ways that are not yet possible in nonhuman primates and will help us identify worthwhile genetic and molecular targets in primates. 2a) Effects of SC suppression on visual detection in mice The SC in one of the most important visual regions in mice, where it is best known for processing visual threats, such as looming stimuli that denote the approach of a predator. In primates the SC plays a pivotal role in detecting behaviorally relevant visual events, such as those used to guide perceptual choices. We have now tested whether mouse SC plays a similar role in visual event detection by transiently inhibiting SC activity during a visual orientation-change detection task. Mice were trained to report visual orientation changes that could occur in either the left or right visual display by licking a central spout to obtain fluid rewards. During this yes/no task, we briefly and unilaterally activated SC intermediate layer GABAergic neurons expressing Channelrhodopsin (ChR2) in vGat-cre mice. We found that SC inhibition induced spatially specific deficits in detection. Optogenetic stimulation caused a significant reduction in hit rates, as well as marked increases in reaction times for near-threshold orientation changes contralateral to the inhibited side. In addition, the behavioral effects caused by SC inhibition were specific to a temporal epoch coincident with known visual responses in the SC. Activating SC GABAergic neurons during a 100-ms visual epoch (starting from 50ms after the orientation change onset) caused a contralateral detection deficit, whereas SC inhibition before or after this epoch did not. Together, these results demonstrate that the mouse SC plays a crucial role in detecting behaviorally relevant visual events that guide appetitive perceptual choices.